Cache Programming for Scientific Loops Using Leases
| Autor: | Benjamin Reber, Matthew Gould, Alexander H. Kneipp, Fangzhou Liu, Ian Prechtl, Chen Ding, Linlin Chen, Dorin Patru |
|---|---|
| Rok vydání: | 2023 |
| Předmět: | |
| Zdroj: | ACM Transactions on Architecture and Code Optimization. |
| ISSN: | 1544-3973 1544-3566 |
| DOI: | 10.1145/3600090 |
| Popis: | Cache management is important in exploiting locality and reducing data movement. This paper studies a new type of programmable cache called the lease cache. By assigning leases, software exerts the primary control on when and how long data stays in the cache. Previous work has shown an optimal solution for an ideal lease cache. This paper develops and evaluates a set of practical solutions for a physical lease cache emulated in FPGA with the full suite of PolyBench benchmarks. Compared to automatic caching, lease programming can further reduce data movement by 10% to over 60% when the data size is 16 times to 3,000 times the cache size, and the techniques in this paper realize over 80% of this potential. Moreover, lease programming can reduce data movement by another 0.8% to 20% after polyhedral locality optimization. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|